Clinical and demographic characteristics of patients

The demographic and clinical characteristics of all 2261 patients are summarized in Table 1 by age group. The patients were divided into four cohorts based on their ages: 18–45, 46–55, 56–65, and 65–80. In the four groups, there are significant differences (P < 0.05) in race, tumor grade, lymph node status, RT, marital status, and local lymphatic biopsy. According to the baseline characteristics of patients with MPTB, the value of RT could be discussed in different cohorts based on different clinicopathological characteristics.

Table 1 Clinical characteristics of MPTB patients diagnosed in 2000–2020 from the SEER databaseSurvival analysis of MPTB patients treated and untreated with adjuvant radiotherapy

The OS and BCSS of MPTB patients treated and untreated with RT were evaluated by the Kaplan–Meier (K–M) survival curves. Compared to patients treated with RT, patients untreated with RT had better OS (P = 0.002) (Fig. 2A) and BCSS (P < 0.001) (Fig. 3A). K–M survival curves revealed there was no statistically significant difference in survival between patients in the age group of 56–65 years and patients in the age group of 66–80 years treated and untreated with RT (age group of 56–65 years: OS P = 0.685, BCSS P = 0.740; age group of 66–80 years: OS P = 0.658, BCSS P = 0.695, respectively). However, for OS and BCSS, there was a statistically significant difference between patients who received RT and patients who did not receive RT in the age groups of 18–45 years and 46–55 years (age group of 18–45 years: OS P = 0.019, BCSS P = 0.016; age group of 46–55 years: OS P < 0.001, BCSS P < 0.001, respectively), and the patients untreated with RT had significant better long-term survival outcomes compared to patients treated with RT. Detailed information on survival analysis results is shown in Figs. 2B–E and 3B–E.

Fig. 2

Kaplan–Meier overall survival curves of MPTB patients in different age groups based on the RT status. A Overall, B 18–45 years, C 46–55 years, D 56–65 years, E 66–80 years

Fig. 3

Kaplan–Meier breast cancer-specific survival curves of MPTB patients in different age groups based on the RT status. A Overall, B 18–45 years, C 46–55 years, D 56–65 years, E 66–80 years

Univariate and multivariate COX regression analyses of different variables

Considering that patients with MPTB of different ages exhibit distinct clinicopathological characteristics, univariate and multivariate Cox regression models were used to assess the effects of variables in four different age groups. After balancing the effects of other factors, as shown in Table 2 and Additional file 1: Table S1, no patients with MPTB were found to be able to benefit from RT, either in younger (age group of 18–45 years: OS HR = 0.694, 95% CI 0.365–1.317, P = 0.263; BCSS HR = 1.097, 95% CI 0.504–2.386, P = 0.816; age group of 46–55 years: OS HR = 0.972, 95% CI 0.635–1.486, P = 0.894, BCSS HR = 1.294, 95% CI 0.748–2.241, P = 0.357, respectively) or older age groups (age group of 56–65 years: OS HR = 1.076, 95%CI: 0.827—1.401, P = 0.585, BCSS HR = 0.633, 95% CI 0.319–1.258, P = 0.192; age group of 66–80 years: OS HR = 0.659, 95% CI 0.377–1.151, P = 0.143, BCSS HR = 1.024, 95% CI 0.483–2.170, P = 0.951, respectively).

Table 2 Multivariate Cox regression model analysis of OS in different age groupsSurvival analysis after propensity score matching

To address variations in baseline characteristics across the four cohorts and minimize bias attributed to other variables, a 1:1 case–control analysis was executed for comparing patients who received radiation therapy (RT) with those who did not receive RT through propensity score matching (PSM). After PSM, eleven factors were enrolled, including the year of diagnosis, race, marital status, laterality, tumor grade, T stage, lymph node status, distant metastatic status, surgery of primary site, local lymphatic biopsy, and chemotherapy (Additional file 2: Table S2). After PSM, no difference in long-term survival outcomes was found in the younger group (age group of 18–45 years: OS P = 0.473, BCSS P = 0.750; age group of 46–55 years: OS P = 0.380, BCSS P = 0.816, respectively) (Figs. 4A, B and 5A, B), although patients untreated with RT before PSM had better long-term survival outcomes than those treated with RT (age group of 18–45 years: OS P = 0.019, BCSS P = 0.016; age group of 46–55 years: OS P < 0.001, BCSS P < 0.001, respectively). Similarly, in the older age groups, there was still no significant difference between patients treated with RT and patients untreated with RT (age group of 56–65 years: OS P = 0.484, BCSS P = 0.290; age group of 66–80 years: OS P = 0.997, BCSS P = 0.763, respectively) (Figs. 4C, D and 5C, D), which is consistent with the pre-PSM results.

Fig. 4

Kaplan–Meier survival estimates of overall survival comparing 1:1 matched Without vs With RT in different age groups. A 18–45 years, B 46–55 years, C 56–65 years, D 66–80 years

Fig. 5

Kaplan–Meier survival estimates of breast cancer-specific survival comparing 1:1 matched Without vs With RT in different age groups. A 18–45 years, B 46–55 years, C 56–65 years, D 66–80 years